Chemistry Unit
Organization of interlaboratory studies and statistical evaluation of analytical data
Type of training
- Group training:
Not available - Individual fellowships:
1 or possibly 2 fellows at a time for not more than 6 months, minimum duration three months
Topics included in the training programme:
- Basic theory of statistical evaluation of data, (including: measures of location and dispersion, methodology of calculation of robust statistical parameters, tests to evaluate the normalcy of data and tests for the identification of outliers).
- Design and organization of interlaboratory comparison studies (proficiency tests and intercomparison exercises) and preparation of relevant documentation.
- Preparation and characterization of test samples to be used in interlaboratory studies (preparation of the test material, assessment of homogeneity and stability of the test material, assignment of the target value, establishment of the analytical uncertainty for the target value).
- Quantifying of uncertainty in the analytical measurements (based on ISO and EURACHEM standards). Calculations using the method of differentiation and the spreadsheet (Kragten) method. Assessment of combined uncertainty using the law of propagation of uncertainty.
- Evaluation of proficiency test data using two different approaches for the validation of analytical results (including: use of the z-Score and u-Score together with two acceptance criteria; accuracy and precision, use and interpretation of Youden plots).
- Validation of individual laboratory results by assessing their accuracy and precision performance.
- Evaluation and interpretation of data obtained through large-scale intercomparison studies. Practical work using a statistical software package (HISTO) developed by AQCS Seibersdorf.
- Introduction to the concept of performing a statistical evaluation in conjunction with the analytical judgement of an experienced analyst.
QA/QC subjects included in the training:
- Introduction to international standards (ISO, EURACHEM, IUPAC, ASTM) relevant to the scope of the fellowship.
Target groups:
Individuals involved in organization of different types of interlaboratory studies (proficiency tests, intercomparison runs, etc.), statistical evaluation and interpretation of data.
Prerequisites:
University degree in chemistry or mathematics or computer technology. Candidates must have a sound understanding of basic statistical principles. At least, three years of practical experience in statistical evaluation of data and organization of interlaboratory studies.
Remarks:
Further details can be obtained from: aqcs@iaea.org